INDEX

Explanations

references to items and their conditions or states

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FailureListener

-0.56

 intensification

-0.55

 getItemId

-0.54

riwal

-0.53

 mutagenicity

-0.51

 pitié

-0.51

 exclusivity

-0.50

 picioare

-0.50

 truncation

-0.50

 possesso

-0.49

POSITIVE LOGITS

 intact

0.83

verwijspagina

0.81

 clean

0.68

 مُعرِّف

0.65

 undamaged

0.64

 properly

0.60

 CreateTagHelper

0.59

 fully

0.59

 securely

0.58

 pristine

0.58

Activations Density 0.399%

references to items and their conditions or states

No Comments

No Known Activations